Data Warehouse Integrations

Last updated: July 10, 2026

We support connecting data warehouses to the Numeric platform to allow joining your internal data against general ledger transactions for reconciliation and flux purposes.

Alongside the credentials for your given warehouse, we recommend sharing the set of tables or queries that contain the information you wish to reconcile against or include for the purposes of flux analysis. This helps us configure reports ready for your review during the Technical Preview.

Why connect a data warehouse?

Numeric pulls transaction-level detail directly from your ERP, but your GL only tells part of the story. Data warehouse integrations let you bring in the rest (examples: subledger detail, internal system records, operational metrics) and join it against GL transactions within Numeric.

This enhances two core workflows:

Reconciliation. Compare GL account balances against source-of-truth data that lives in your warehouse. If your team maintains subledger records, billing system exports, payroll registers, or revenue schedules in a warehouse, you can pull those directly into Numeric to reconcile against the corresponding GL accounts -- no more exporting CSVs or toggling between systems.

Flux analysis. Surface richer context behind period-over-period variances. When Numeric's AI flux writer drafts variance explanations, it can reference warehouse data alongside your GL transactions; think payroll breakdowns by department, marketing spend by campaign, revenue detail by customer, or headcount changes.

BigQuery

We support BigQuery's Service Account authentication.

  1. Create a Service Account within the GCP console.

    1. Service account name: Numeric Reader

    2. Description: Account for numeric.io to read BigQuery tables.

  2. Add roles:

    1. BigQuery Job User

    2. BigQuery Data Viewer

      1. This role can be granted to specific datasets and table within BigQuery to limit what Numeric Reader can read from.

    3. BigQuery Metadata Viewer

  3. Create a key for this service account and download it as JSON. Share this key with us.

Snowflake

We support Snowflake's username plus key pair authentication. This requires running a set of SQL commands to create the appropriate user and permissions:

CREATE USER IF NOT EXISTS NUMERIC LOGIN_NAME = 'NUMERIC';
CREATE ROLE IF NOT EXISTS NUMERIC_READER;

GRANT ROLE NUMERIC_READER TO USER NUMERIC;

-- Give the NUMERIC user access to your relevant warehouse.
GRANT USAGE ON WAREHOUSE <WAREHOUSE_NAME> TO ROLE NUMERIC_READER;

-- Give the NUMERIC user access to your relevant databases.
GRANT USAGE ON DATABASE <DATABASE_NAME> TO ROLE NUMERIC_READER;

-- Give the NUMERIC user access to your relevant schemas.
GRANT USAGE ON SCHEMA <SCHEMA_NAME> TO ROLE NUMERIC_READER;

-- Allow the NUMERIC user to run SELECT on the each database and schema 
-- specified above.
GRANT SELECT ON ALL TABLES IN SCHEMA <DATABASE_NAME>.<SCHEMA_NAME> TO ROLE NUMERIC_READER;

As a last step, you need to add an RSA public key to the NUMERIC user. Please reach out to support@numeric.io with the following information, which allows us to generated this public key:

  1. Snowflake account URL, e.g. https://<account>.snowflakecomputing.com

  2. Snowflake warehouse name, e.g. COMPUTE_WH

  3. Snowflake username, only if different from NUMERIC.

Finally, you need to run the below command to set the RSA public key we share with you:

ALTER USER NUMERIC SET RSA_PUBLIC_KEY='<PROVIDED BY NUMERIC SUPPORT>';

Databricks

We support Databricks' OAuth 2.0 machine-to-machine authentication. Follow the instructions in Databrick's documentation, then share with us:

  1. workspace_url

  2. client_id

  3. client_secret

  4. warehouse_id

Please ensure that the service principal attached to these credentials only has read access to the tables you wish to leverage within the Numeric platform.

NetSuite Saved Searches and SuiteQL

Numeric can reuse the same credentials we already store to sync over NetSuite transactions onto the platform. We recommend adding any additional permissions required to read specific saved searches or SuiteQL tables to these same credentials. 

This is especially useful for teams that already have saved searches built for internal reporting: items like revenue waterfall detail, custom transaction searches by subsidiary, or intercompany elimination schedules. Rather than rebuilding that logic, you can point Numeric at the same searches your team already relies on and use that data for reconciliation or flux analysis directly in the platform.

Workday

We support Workday's WQL interface via their API Client credentials.

  1. Create a new API Client for Integrations.

  2. Ensure Non-Expiring Refresh Tokens is enabled.

  3. Share with us:

    1. The endpoint URL, e.g. https://impl-services1.wd12.myworkday.com

    2. Your tenant name

    3. Client ID

    4. Client Secret

    5. Refresh Token

Workday's WQL interface gives Numeric access to query structured data across your Workday tenant. Common use cases include pulling payroll summary data, headcount by cost center, or compensation details to reconcile against payroll-related GL accounts or to enrich flux explanations for personnel-related variance.